When a child experiences emotional and behavioral problems that interfere with normal functioning, a psychological assessment is often requested. As a typical first step in the assessment process, a psychologist would have a teacher or parent complete a standardized child behavior checklist. In a thorough assessment, this would only be the beginning, and a more detailed analysis of the problem behavior would follow. Regardless of what follows, however, the scale scores that are derived from the standardized checklist may influence how a clinician initially conceptualizes a child's difficulties and may become part of a child's clinical file or subsequent written assessment.

Our recent research on standardized child behavior checklists suggests that they are problematic because they systematically obscure information about the social context in which children's behavior occurs. Important information about the meaning of the child's behavior problems, and situational factors contributing to the child's difficulties, are lost, potentially contributing to the conclusion that the behavior problem is "in the child" rather than being a product of both the child's behavior and the surrounding social environment.

Child behavior checklists are widely used in both research and clinical practice, and therefore may have a broad influence on how people think about child psychopathology. Publicity for one of the most popular of these checklists indicates that it has been used in more than 2,400 studies by more than 4,500 investigations.

Although the details of the different child behavior checklists vary, in essence the teacher or parent is asked to indicate how often a child displays a number of problem behaviors over a period of several months. For example, the respondent might be asked to summarize how often the child cries, whines, hits or threatens other people. Using standardized scoring procedures, the ratings for related items are averaged to form "scale scores," (e.g., aggression, withdrawal), which are then compared to national norms based on large samples of children.

Authors of these inventories suggest that filtering out situational variation is one of their major strengths. However, others suggest that this is one of their greatest weaknesses and prefer to emphasize the complex combination of child and environment factors that can contribute to child psychopathology.

Context adds understanding

An example illustrates one of the main concerns raised about child behavior checklists from this contextual point of view:

Kyle and Sheryl are similar in their overall frequency of displaying aggression. However, when we assess context, we learn that Kyle is aggressive when threatened, whereas Sheryl is aggressive when talked to in a friendly way.

Although the causal factors related to these children's aggressive behavior are likely to be different, and different interventions probably would be needed, aggression scale scores do not detect this patterning difference and could indicate that Kyle and Sheryl are equally aggressive.

The problem illustrated by this hypothetical example was demonstrated in a recent study conducted in a summer residential treatment program for children with emotional and behavioral problems.

Children were divided into commonly studied clinical groups using a popular child behavior checklist. These groups were formed by identifying those children who were high or low on two broad behavior problem scales: "Internalizing" (a composite score of withdrawal, anxiety/depression, and somatic complaints) and "Externalizing" (a composite score of aggressive and delinquent behavior).

Externalizing children were high on externalizing only, Internalizing children were high on internalizing only, and Mixed (or "comorbid") children were high on both. The latter group is of special interest because they are at heightened risk for subsequent psychological problems. Based on their scale scores alone, one might expect that Mixed children simply display high rates of externalizing behavior (like the Externalizers) and high rates of internalizing behavior (like the Internalizers). However, when we examined how these groups of children responded in different social situations, we found important differences in the contexts that elicited their aggressive or withdrawn behavior, differences that were obscured by their similar scale scores.

Externalizing children showed a specific pattern of elevated aggression in response to aversive events (e.g., adult warning). Internalizing children showed a specific pattern of elevated withdrawal in response to aversive peer events (e.g., peer tease). Mixed children did not display either a blend or an average of these patterns for the two other groups; instead, they displayed a very distinctive pattern that consisted of elevated rates of aggression and withdrawn responses to positive events (e.g., friendly peer talk, adult praise). Thus, Mixed children were especially distinctive in the contexts in which they displayed their problem behavior, becoming aggressive and withdrawn in non-aversive situations, where the other groups were neither aggressive nor withdrawn.

Summarizing a child's behavior broadly over contexts obscures such information about the contextualized patterning of children's behavior, information that may be essential to the therapist, teacher or parent in designing effective interventions.

Differences appear same

A second hypothetical example highlights another kind of confusion that can result when the social context of a child's behavior problems is ignored:

Brandon and Kyle are similar in their overall frequency of displaying aggression. However, examination of contextual factors indicates that Brandon is teased frequently by peers, but is only occasionally aggressive when teased. Kyle, on the other hand, is rarely teased by peers, but is frequently aggressive when teasing does occur.

Intuitively and clinically, these children are fundamentally different: Brandon's aggressiveness appears more related to his hostile peer group, whereas Kyle's aggression is more related to how he responds to events. Nevertheless, the two children's checklist scale scores for aggression could be equivalent.

In a recent study we conducted with Kristen Lindgren in a special education setting, children who fit the descriptions of these hypothetical children were identified. Thus, the two groups differed both in their social environments and in how they responded: One group was frequently provoked, but was unlikely to respond aggressively when this occurred; the other group was provoked infrequently, but when this did occur they were likely to respond aggressively.

As expected, when teachers were asked to rate them using a standardized checklist, the checklist scale scores could not discriminate between these fundamentally different groups of children.

These findings were extended in a follow-up experiment in which participants read descriptions of fictitious children that resembled Brandon and Kyle. Interestingly, participants were quite sensitive to the differences between these targets in their social environments and their responses when we asked them directly. In some, rather than tapping people's knowledge of contextual factors, standardized instruments apparently lead them to ignore this information in order to produce simple (but contextually impoverished) scale scores.

As third-party reimbursement for comprehensive psychological evaluation decreases, it is likely that clinicians will be encouraged to rely on brief and inexpensive methods of assessment such as child behavior checklists.

There are several dangers associated with reliance on such methods. First, information about the social environment that is needed to interpret a child's behavior may be obscured. Second, this neglect of contextual factors creates the potential for misguided interventions that are based on measures of behavioral output rather than environmental factors that may contribute to the child's difficulties. Third, reliance on problem behavior scale scores could interfere with communication between the clinician and the teacher or parent who attend naturally to the contextual factors influencing a child's behavior problems. Fourth, child behavior checklists may reinforce the view that the problem or "syndrome" resides in the child.

In the short run, care should be taken to obtain the detailed contextual information that gives meaning to a child's behavior problems in follow-up interviews, and caution should be used before attributing the problems score to the child's individual difficulties.

In the long run, it will be important to develop a new generation of child behavior checklists that focus on child behavior problems and their social contexts.

Audrey Zakriski is an assistant professor of psychology at Connecticut College, alzak@conncoll.edu, and a licensed clinical psychologist in Rhode Island. Jack Wright is an associate professor of psychology at Brown University.

Recommended Reading:

Achenbach TM: Empirically based taxonomy: How to use syndromes and profile types derived from the CBCL/4-18, TRF, & YSR. Burlington, VT: University of Vermont Department of Psychiatry. 1993.

Wright JC, Zakriski AL, Drinkwater M: Developmental psychopathology and the reciprocal patterning of behavior and environment. Journal of Consulting and Clinical Psychology 1999; 67, 95-107.

Wright JC, Lindgren K, and Zakriski AL: Syndromal versus contextually sensitive assessment of child psychopathology: differentiating environmental and dispositional determinants of behavior. 1999. Manuscript under editorial review.

In the past two years in America, nearly 400,000 children joined the ranks of the extremely poor. Now there are 6.4 million children in families struggling to get by on incomes less than one-half of the poverty line. For a family of three, that's less than $6,400 a year. Think about what it must be like to try to feed, clothe and raise a family in the United States on such a small amount of money.

The bad news doesn't end there either. The number of poor children, 14.1 million, remains higher than it was at any time between 1966 and 1990. And, in 1997, a record number of working families with children were living in poverty.

Over the past few years, children in low-wage working families have been bearing the brunt of rising poverty, falling incomes, vanishing employer-paid health benefits, and a weakening safety net. When children grow up in poverty, they are more likely to perform poorly in school and are less likely to finish high school, let alone college. They are less likely to succeed as adults in earning a living wage or live up to their full potential for contributing to the rest of society.

There is some good news in the middle of all this despair. The American people want to help and they want to help now, according to a survey by the W. K. Kellogg Foundation. This poll clearly shows the American people think it's time for working families to get the support they need to provide for their children.

More than eight out of 10 Americans said the working poor, along with those who are trying to go from welfare to work, should be given help to raise their children out of poverty. The poll shows there is overwhelming support for assisting all low-income families with health coverage, quality child care, job training and more. It also shows that Americans want the federal and state governments to play a leading role in providing these supports.

For too long there's been a chasm between what the American people want and what they get from their leaders. It's grown bigger and bigger, just like the chasm dividing the rich from the poor in this nation. And it's our children who are falling into this abyss.

We must build the spiritual and political will to get children what they need to grow up healthy and ready to succeed. This is the time to make sure every child is given health coverage. This is the time to make sure every working family has access to affordable, quality child care if they need it.

At the same time we advocate for services that aren't available yet, we must also maintain strategic and sustained outreach to let working families know about the services that are out there for the, such as the state Children's Health Insurance Program (CHIP).

As Congress and the President debate how to use our remarkable budget surplus and as state legislators come together in the light of a very strong economy, they must consider the will of the American people. This nation needs to put children at the real, not the rhetorical, center of its priorities.

Reprinted with permission from The Children's Defense Fund CDF Reports, March 1999, p. 3.

Recommended Reading:Expanded section